Reno

Inspiration Home renovation is exciting, but turning an idea into an actionable plan is overwhelming. Homeowners and house flippers often have to coordinate designers, contractors, suppliers, pricing, and delivery across multiple platforms. We created Reno to bring that fragmented process into one AI-powered experience. What it does Reno allows users to upload photos of one room or an entire house and describe their renovation vision. It generates renovated concepts while maintaining a consistent style across rooms. After the user selects a design, Reno creates a tailored material plan with paint, flooring, lighting, fixtures, finishes, quantities, and product specifications. Reno can also compare retailer prices based on the project’s location, link users directly to product sources, organize materials into a cart, and prepare the order for checkout. It finds nearby contractors and displays clearly labeled preliminary labor scenarios until verified estimates are received. How we built it We built Reno as a full-stack web application using React, Next.js, TypeScript, and Cloudflare infrastructure. OpenAI’s image and language models power the renovation concepts, visual material analysis, structured product recommendations, retailer research, and contractor discovery. The application processes uploaded room photos, creates design variations, converts the selected design into structured material data, and searches current retailer listings for comparable products. Location data helps personalize stores, pickup information, delivery availability, and nearby contractor results. Challenges we ran into One major challenge was preserving the structure of the uploaded room while changing its finishes and overall style. We also needed to keep designs visually consistent when users uploaded multiple rooms. Turning a generated image into a reliable shopping list was another difficult problem. Product availability, pricing, package sizes, and store locations constantly change. Some retailers also restrict access to product images or do not publish prices consistently. We addressed this by linking directly to source pages, searching for close in-stock alternatives, and clearly identifying information that could not be verified. We also had to distinguish between preliminary labor scenarios and genuine contractor quotes so users would not mistake an AI-generated estimate for a confirmed bid. Accomplishments that we're proud of We are proud that Reno goes beyond generating attractive inspiration images. It connects the creative phase of renovation with practical planning, sourcing, budgeting, and checkout. The platform supports up to 30 room photos, helps maintain a unified whole-home design language, produces design-specific material plans, compares local retailer prices, and identifies nearby contractors. We also built safeguards that distinguish verified information from estimates and prevent misleading calculations such as negative savings. What we learned We learned that generating a beautiful room is only the beginning. A useful renovation assistant must understand product specifications, local availability, quantities, budgets, logistics, and the difference between inspiration and verified purchasing information. We also learned that transparency is essential in an AI-powered marketplace. Users need to know where product information came from, when a price was checked, and whether a labor amount is illustrative or contractor-confirmed. What's next for Reno Next, we want to add verified contractor outreach through phone, text, and email; collect comparable written bids; and replace preliminary labor scenarios with confirmed estimates. We also plan to add automatic room measurements, more accurate quantity calculations, retailer inventory integrations, delivery scheduling, secure multi-store payments, order tracking, contractor booking, shared project workspaces, and tools for managing an entire renovation from the first photo to the final walkthrough.

Built With

• codex
• openai